Documente Academic
Documente Profesional
Documente Cultură
Abbreviations
fMRI
functional magnetic resonance imaging
Evidence that these two memory systems are independent comes from patients with focal lesions to the amygdala or hippocampus. In a classic fear conditioning
paradigm, during which a neutral blue square is paired
with an aversive shock to the wrist, patients with amygdala damage fail to show a normal physiological fear
response to the blue square, even though they are able
to report that the blue square predicted the shock [2].
Patients with damage to the hippocampus show the
opposite pattern [3]. That is, they demonstrate a physiological arousal response to the blue square, but are not
able to consciously recollect that it was paired with the
shock. This double dissociation highlights the independent functions of these two memory systems. Even
though these two systems can operate independently,
they also interact in subtle and important ways. In this
review, I outline how the amygdala and hippocampal
complex influence each other when emotion and memory
come together.
Introduction
One of the primary advances in the study of memory over
the past half century is the growing recognition that there
are multiple memory systems that are governed by distinct and interacting neural substrates [1]. Investigations
examining the influence of emotion on memory have
primarily focused on two medial temporal lobe memory
systems (see Figure 1). The first is linked to the amygdala
and is more or less specialized for the processing of
emotion. The hallmark of this memory system is that
it is crucial for the acquisition and expression of fear
conditioning, in which a neutral stimulus acquires aversive properties by virtue of being paired with an aversive
event. The second is linked to the hippocampal complex
and is necessary for declarative or episodic memory. This
memory system can be thought of as a primary memory
system in humans, in that it governs the function most
often referred to as memory, that is, the recollection of
events at will.
Current Opinion in Neurobiology 2004, 14:198202
Figure 1
patients with varying degrees of pathology to the hippocampus and amygdala during the encoding of emotional
and neutral words. Using fMRI, they found that greater
left amygdala pathology predicted both worse subsequent
memory for the emotional words and less activity in the
left hippocampus. Memory for neutral words was only
related to the degree of hippocampal pathology. Interestingly, the relationship between the pathology of the
amygdala and hippocampus and the activity in response
to emotion words was bi-directional. More left hippocampal pathology predicted less activity in the left amygdala
and more activity in the right amygdala, which suggests a
mutual dependence of the hippocampus and amygdala
when remembering emotional stimuli.
The brain imaging and patient studies cited above support a role for the human amygdala in modulating the
hippocampal complex, but do not indicate that this modulation alters the consolidation or storage of memory per
se. This question was addressed in two recent studies by
Cahill and co-workers. Using a pharmacological [25] and
pain [26] manipulation that elicited a stress hormone
response immediately after a target stimulus was
encoded, they were able to demonstrate enhanced memory for this stimulus. These results strongly support the
conclusion that emotion can alter the retention of emotional events and are consistent with the animal models
suggesting a role for the amygdala in the modulation of
hippocampal consolidation.
An emerging topic in our efforts to understand the
mechanisms underlying the amygdalas influence on
emotional memory is the unique roles of the left and
right amygdala. Recent brain imaging studies have suggested that the left and right amygdala could be differentially involved in memory for emotional stimuli
depending on the sex of the subject. Specifically, two
recent studies have shown that the left amygdala is
correlated with later memory for emotional stimuli in
female subjects, whereas the right amygdala is correlated
with memory for emotional stimuli in male subjects
[27,28]. It is unclear if this laterality difference could
be related to sex differences in stimulus processing strategies or other factors. However, studies examining emotional memory or physiological responses to emotional
stimuli in patients with amygdala damage have failed to
find such sex differences. These studies have tended to
be consistent with previous studies on hippocampal function showing a material specific involvement of the left
and right amygdala for verbal and visual material, respectively [19,29,30]. It is unclear at this time how data from
these two different techniques will inform our understanding of the specific roles of the left and right amygdala
in episodic memory for emotional events.
In addition to its primary role in the acquisition and
physiological expression of conditioned fears, the amygCurrent Opinion in Neurobiology 2004, 14:198202
Figure 3
(a)
(b)
Amygdala activation to an episodic representation of fear. Left amygdala activation to instructed fear. Composite activation response to (a) threat
versus safe stimuli and (b) selected individual subjects. Adapted from Phelps et al. [31].
subjects to use strategies to alter their response to emotional stimuli. A recent fMRI study instructed subjects to
reappraise the emotional significance of a negative scene
by trying to interpret the events depicted in a nonemotional or positive light [32]. Both the acquisition
and the appropriate application of this strategy require
hippocampal-dependent memory. The reappraisal strategy was successful in diminishing both the reported
emotional reaction to the negative scenes and the amygdala response.
The recollection of emotional stimulus properties and
strategies acquired through instruction requires the formation of hippocampal-dependent memories. These episodic memories can influence our emotional reactions, at
least in part by modulating the amygdala.
Conclusions
The amygdala and hippocampal complex govern two
independent memory systems that interact when emotion
meets memory. We are just beginning to understand the
subtleties of these interactions in humans and there are
still several unanswered questions.
Although we know more about the mechanisms that
underlie the amygdalas influence on hippocampaldependent episodic memory, it is not clear how the
modulation of attention and encoding complements the
modulation of consolidation. It has been suggested that
the amygdala primarily enhances episodic memory for the
gist of an emotional event at the expense of details [34],
which may be related to attention. It has also been
suggested that this gist versus peripheral detail memory
enhancement might interact with sex and laterality when
memories are consolidated [35]. Researchers are only
beginning to investigate these topics and it is unclear
which, if any, of these factors will prove to be important.
www.sciencedirect.com
Acknowledgements
The author would like to acknowledge K Nearing and B Sedgewick
for assistance with figure preparation and M Delgado for helpful
comments.
Current Opinion in Neurobiology 2004, 14:198202
2.
3.
4.
5.
6.
7.
8.
31. Phelps EA, OConnor KJ, Gatenby JC, Grillon C, Gore JC, Davis M:
Activation of the left amygdala to a cognitive representation of
fear. Nat Neurosci 2001, 4:437-441.
32. Oschner KN, Bunge SA, Gross JJ, Gabrieli JDE: Rethinking
feelings: an fMRI study of the cognitive regulation of emotion.
J Cogn Neurosci 2002, 14:1215-1229.
This study examines the neural substrates underlying our ability to
regulate our emotional responses and demonstrates that the amygdala
response might be altered by cognitive strategies.
33. Schaefer SM, Jackson DC, Davidson RJ, Kimberg DY,
Thompson-Schill SL: Modulation of amygdalar activity by the
conscious regulation of negative emotion. J Cogn Neurosci
2002, 14:913-921.
34. Adolphs R, Denberg NL, Tranel D: The amygdalas role in
long-term declarative memory for gist and detail.
Behav Neurosci 2001, 115:983-992.
35. Cahill L, van Stegeren A: Sex-related impairment of memory
for emotional events with beta-adrenergic blockade.
Neurobiol Learn Mem 2003, 79:81-88.
www.sciencedirect.com